Journal of Economics Bibliography

www.kspjournals.org
March 2016

Volume 3

Issue 1

The Effect of Exports and Imports on Economic

Growth in the Arab Countries: A Panel Data
Approach
By Bader S.S. HAMDANa
Abstract. The study focused on the effect exports and imports on economic growth in the
Arab countries during the period 1995 to 2013. The study used panel data approach in 17
countries: (Jordan, United Arab Emirates, Bahrain, Tunisia, Algeria, Saudi Arabia, Sudan,
Oman, Qatar, Kuwait, Lebanon, Egypt, Djibouti, Mauritania, Morocco, Yemen and
Palestine). The study used panel data approach by E views program. The study found that
the effect exports and imports have positive effect of economic growth in the Arab
countries during the period 1995 to 2013. The study recommended it is important indicator
for measuring the efficiency and effectiveness of the work element in achieving a certain
level of the output in the production process. There is need to increase the imports of
technology for increasing labor productivity which can directly promote economic growth,
and thus improve the standards of living in the Arab countries.
Keywords. Panel Data Approach, Housman, Exports, Imports.
JEL. E62, H54, O40.

1. Introduction

he exports and imports play important role on economic growth in the

developed and developing countries, economic growth is one of the most
important determinants of economic welfare. The relationship between
exports and economic growth is a frequent topic of discussion, when economists
try to explain the different levels of economic growth between countries. Exports
of goods and services represent one of the most important sources of foreign
exchange income that ease the pressure on the balance of payments and create
employment opportunities. According to Feder (1982), exports contribute to
economic growth in avariety of ways - greater capacity utilization, economies of
scale, incentives for technological improvement and pressure of foreign
competition, leading to more efficient management. Thus, marginal factor
productivities are expected to be higher in export industries than in non-export
industries. Thecross-sectional analysis by Feder (1982) and Ram (1987) confirm
this productivity differential for developing countries, although the differential
coefficients in Feder (1982) for developed countries are insignificant.The exports
growth in the Arab countries was 14.78 percent per annum during the period 19952013. The average export was USD 667.55 billion (47.87 percent of GDP) during
the period 1955 to 2013.The exports in Arab countries increased from USD 178.58
billion (34.34 percent of GDP) in 1995 to USD 216.26 billion (35.49 percent of
GDP) in 1997, then the exports in Arab countries decreased to USD 172.18 billion
a

Department of Economy, Al-Azhar University, Egypt.

. 0537 27 27 50
. bsshamdan@hotmail.com

Journal of Economics Bibliography

(29.30 percent of GDP) in 1998.The exports in Arab countries decreased to 877.91
billion (48.94 percent of GDP) in 2009. The exports in Arab countries was
constantly rising during the period 2010 to 2013, where the exports in Arab
countries increased from 1077.15 billion (51.34 percent of GDP) in 2010 to
1558.60 billion (54.81 percent of GDP) in 2013.
The imports growth in the Arab countries was 13.38 percent per annum during
1995-2013. The average imports were USD 508.06 billion (35.76 percent of GDP)
during period 1955 to 2013.The imports in Arab countries increased from USD
173.92 billion (33.44 percent of GDP) in 1995 to USD 187.86 billion (31.97
percent of GDP) in 1998 and then decreased to USD 180.57 billion (28.21 percent
of GDP) in 1999. During the period 2003 to 2008 the imports in Arab countries
increased from USD 270.28 billion (33.04 percent of GDP) in 2003 to USD 880.82
billion (42.47 percent of GDP) in 2008, then to USD 1130.94 billion (41.19 percent
of GDP) in 2012. The Arab countries recorded imports of USD 1226.76 billion
(43.14 percent of GDP) in 2013.
This paper focused to estimate the effect exports and imports on economic
growth in the Arab countries during the period 1995 to2013, by panel data method.

2. Literature Review
Several studies address the importance of exports and imports on economic
growth. The findings of these studies indicate that exports and imports have a
statistically significant positive impact on economic growth. We can summarize
some of these studies that have addressed the issue of effect exports and imports on
economic growth as follows:
Afaf, (2015) investigated the impact of exports and imports on the economic
growth of Tunisia over the period1977-2012. The study used Granger Causality
and Johansen Cointegration approach for long run relationship using Augmented
Dickey-Fuller (ADF) and Phillip-Perron (PP) stationarity test, the variable proved
to be integrated of the order one (1) at first difference. Johansen and Juselius
Cointegration test was used to determine the presence or otherwise of a
cointegrating vector in the variables. The study found finding is clarified that
export, import and GDP are found of order one (t) stationary at the first differences.
Therefore, the variables were found to be integrated of order one. The cointegration
test confirmed that GDP, export and import are cointegrated, indicating an
existence of long run equilibrium relationship between all the variables.
Abugamea, (2015)Examined the both the long run and short run relationships
between economic growth, exports and imports of Palestine for the time period
1968-2012, The study used the cointegration and Granger causality tests. The study
found results based on vector error correction models (VECM) confirm the
existence of a long run relation between imports and economic growth and show
that both exports and imports are the main determinants of economic growth in the
Palestinian case. Causality tests confirm VECM results that imports cause changes
in economic growth in the long run but not in the short run. Kalaitzi (2013)
examined the relationship between exports and economic growth in the United
Arab Emirates over the period 1980-2010. The study applied the two-step EngleGranger cointegration test and the Johansen cointegration technique in order to
confirm or not the existence of a long-run relationship between the variables.
Moreover, this study applied a Vector Autoregression Model in order to construct
the Impulse Response Function and the Granger causality test to examine the
causality between exports and economic growth. The findings ofthis study
confirmed the existence of a long-run relationship between manufactured exports,
primary exports and economic growth. In addition, the Granger causality test
JEB, 3(1), B.S.S. Hamdan. p.100-107.
101

Journal of Economics Bibliography

showed unidirectional causality between manufactured exports and economic
growth. Thus, further increase in the degree of export diversification from oil could
accelerate economic growth in UAE. Elbeydi (2010) investigated the relationships
between export and economic growth in Libya. An econometric model has been
developed and estimated in order to determine the direction of causality in both,
short and long run. The annual time series used for the estimation cover the time
period 1980 2007. The findings indicate that the income, exports and relative
prices are cointegrated. The long run bidirectional causality between the exports
and income growth has been also proved. The study result indicates that the export
promotion policy contributes to the economic growth in Libya. Al-Swaee (2008)
estimated the role exports in economic growth in the west Asia region (oil and nonoil countries) during the period 1993 to 2003. This study used a panel data, the
study found that the productivity of exports is positive effect on economic growth
in oil countries and negative in the non- oil countries. The study recommended the
adoption the policy of export- oriented to benefit from the comparative advantage
in export of goods that the local resources a variable in all states of the region.

3. Methodology and Data

Based on the foregoing explained in the previous chapters, using a variety of
applied studies for different models in estimating the determinants of economic
growth in addition to the use of different methodologies, accordingly, the standard
model in this study, the general equation is the following:
= , , ,
Thus, our growth function becomes;
= C + 1 EX + 2 IM + 3 GCF + 4 LO + t
Where:
: Economic growth (proxy for Gross domestic product in period t,
(current price USD)
EX : Export of goods and services in period t, (current price USD)
IM : Import of goods and services in period t, (current price USD)
GCF : Gross capital formation in period t, (current price USD)
LO : Labor force
C: Constant
t : The standard error
By taking the log of both sides of the equation becomes:
= C + 2 logEx + 3 logIm + 4 GCF + 5 loglo + t
Data have been collected during the period 1995 to 2013, for17 countries in
Arab countries :( Jordan, united Arab Emirates, Bahrain, Tunisia, Algeria, Saudi
Arabia, Sudan, Oman, Qatar, Kuwait, Lebanon, Egypt, Djibouti, Mauritania,
Morocco, Yemen and Palestine). Number of countries which could have been part
of the sample were omitted due to lack of sufficient data on some of the variables
under investigation because of the unstable political the situation. The sample
under study the required secondary Data was collected from official sources like
World Bank data.

4. Method
The study used the panel data method, through the use of three models is: Poold
regression model (PRM), fixed effect model (FEM) and random effect model
(REM). To know any better models to be used in the analysis will be applied tow
test: the first test (test LM) Lagrange multiplier proposal from Preusch and Pagan
in (1980). This test is used to choose between (PRM), (FEM) or (REM), the second
test, Housman test (1978), for choose between (FEM), (REM).
JEB, 3(1), B.S.S. Hamdan. p.100-107.
102

Journal of Economics Bibliography

4.1. The Pooled Effect Model
It can clarify the compound regression model as follows:
= + +

(1)

Suppose pooled regression model homogeneity of variances random error

between the countries under study limits ( 2 = 2 ), together with zero
covariances between countries Cov it , = 0 for i j. (Alexiou, 2001:
p.6).The model also assumes forming Fixed limit transactions (, ) and slope
coefficients (, ) for all countries.
Table 1.Results Pooled Model
Variable
Coefficient
Std. Error
LOG(EX)
-0.925845
0.432343
LOG(IM)
2.215917
0.580230
LOG(GCF)
1.865242
0.514735
LOG(LO)
-0.090511
0.224829
C
17.20529
3.181495
Note:R-Square = /0.641896 a adjusted R-Square = 0.637392

t-Statistic
-2.141457
3.819033
3.623696
-0.402577
5.407926

Prob.
0.0330
0.0002
0.0003
0.6875
0.0000

As shown table (1) the independent variable (export, import and gross capital
formation) was significant at level of 1%, the labour was in significant at level of
1%. The exports and labour had negative effect, on economic growth in the Arab
countries, import and gross capital formation was positive effect on the economic
growth in the Arab countries. Also the R-Square reached 0.637 in the pooled effect
model.

4.2. The Fixed Effect Model

The fixed effects model is simply a linear regression model in which the
intercept terms vary over the individual units i, (Dinardo, Johnston, 1997:p.397).
Yit = 1 1it + 2 2it + + X it + it

(2)

Where it is usually assumed that all are independent of all , we can write
this in the usual regression framework by including a dummy variable for each unit
in the model (Hsiao, 2003:p.96). That is,
yit =

N
j=1 j dij

+ xit + it

(3)

Where = 1 if i=j and 0 elsewhere. We thus have a set of N dummy variable

in the model. The parameters 1 . . , and can be estimated by ordinary
least squares in (3). The implied estimator for is referred to as the Least Squares
Dummy Variable (LSDV) estimator. It may, however, be numerically unattractive
to have a regression model with so many repressors.
Table 2.Results Fixed Effect Model
Variable
Coefficient
Std. Error
LOG(EX)
0.306909
0.042141
LOG(IM)
0.258478
0.053805
LOG(GCF)
0.123597
0.034310
LOG(LO)
0.373511
0.062443
C
15.31687
0.857984
Note:R-Square = 0.999/ a adjusted R-Square =0.999

t-Statistic
7.282905
4.803968
3.602402
5.981660
17.85217

Prob.
0.0000
0.0000
0.0004
0.0000
0.0000

JEB, 3(1), B.S.S. Hamdan. p.100-107.

103

Journal of Economics Bibliography

As shown in table (2) the independent variables export, import, gross capital
formation and labour was significant at level of 1%, also all the independent
variables. Also the R-Square reached 0.999 in the pooled effect model.

4.3. The Lagrange Multiplier (LM) Test:

The Lagrange Multiplier model is as follows (Greene, 2002:p.299)
LM = 2

NT
T1

2
N
T
i=1 t =1 it
N
T
i=1 t =1 it

~x 2

(4)

If the value of (p- value) statistical test (LM),is statistically significant for this
test, it means that FEM, REM, would be better than PRM. It this value is not
statistically significant for the same test, this means that PRM will be better than
the FEM, REM.
Table 3.Result LM Test
Effects Test
Cross-section F
Cross-section Chi-square

Statistic
10742.605807
2049.722519

d.f.
(16,302)
16

Prob.
0.0000
0.0000

As shown table (3) the effects models better than the pooled model.

4.4. The Random Effect Model

It is commonly assumed in regression analysis that all factors that affect the
dependent variable but that have not been included as repressors can be
appropriately summarized by a random error term. In our case, this leads to the
assumption that the are random factors, independently and identically distributed
over individual distributed over individuals. Thus we write the Random Effects
Model as,
yit = + xit + i + it , it IID o, 2 ; i IID o, 2

(5)

where + is treated as an error term consisting of two components: an

individual specific component, that this not vary over time, and a remainder
components, That is assumed to be uncorrelated over time, this is all correlation of
the error terms over time is attributed to the individual effects . It is assumed
that and are mutually independent and independent of (for all j and s).
This implies that the OLS estimator for and from (5) is unbiased and
consistent. The error components structure implies that the composite error term
+ exhibits a particular form of autocorrelation (unless 2 = 0 )
(Verbeek,2000).
Table4.Results Random Effect Model
Variable
Coefficient
LOG(EX)
0.306790
LOG(IM)
0.257684
LOG(GCF)
0.124280
LOG(LO)
0.376878
C
15.26865
Effects Specification
Cross-section random
Idiosyncratic random
R-Square
A adjustedR-Square

Std. Error
0.042116
0.053786
0.034305
0.062245
1.209722

t-Statistic
7.284369
4.790881
3.622762
6.054760
12.62163
S.D.
3.527444
0.150530

Prob.
0.0000
0.0000
0.0003
0.0000
0.0000
Rho
0.9982
0.0018
0.931195
0.930330

JEB, 3(1), B.S.S. Hamdan. p.100-107.

104

Journal of Economics Bibliography

As shown in table (4) the independent variable (export, import, gross capital
formation and labour were significant at level of 1%.The exports, imports, gross
capital formation and labour were positive effect on economic growth in the Arab
countries. Also the R-Square reached 0.931 in the random effect model.

4.5. The Hausman Test

Hausman test is used decide between Fixed Effect model and Random Effects
model. Null hypothesis is that the preferred model is Random Effects Model vs. the
alternative is the Fixed Effects model. It basically tests whether the unique errors
(ui) are correlated with the regresses; the null hypothesis is they are not,
(Chmelarova, 2007).
Table 5.Result Hausman Test
Test Summary
Chi-Sq. Statistic
Cross-section random
19.148622
Cross-section random effects test comparisons:
Variable
Fixed
Random
LOG(EX)
0.306909
0.306790
LOG(IM)
0.258478
0.257684
LOG(GCF)
0.123597
0.124280
LOG(LO)
0.373511
0.376878

Chi-Sq. d.f.
4

Prob.
0.0007

Var(Diff.)
0.000002
0.000002
0.000000
0.000025

Prob.
0.9343
0.5758
0.2164
0.4977

As shown in table (5) the fixed effects models better than the random effects
model.So they study was analysed the results fixed effects models:
= 15.31687 + 0.306909Ex + 0.258478 Im + 0.123597 GCF
+ 0.373511 LO

5. Results and Discussion

The study found the all independent variable had significant at level 1%. The
study also found that the exports, imports, gross capital formation and labor had
positive effect on economic growth in the Arab countries during the period 1995 to
2013.The coefficient of determination R2 is 0.999which means that the explanatory
variables explained a total variation of 99 percent of the dependent variable
(GDP).The exports were a significant at the level of 1% and positive effect on
economic growth in the Arab countries. Also the elasticity of exports in the Arab
countries during the study period reached 0.30%, if the exports increased by 100%
in the Arab countries the economic growth increased by 30 percent. The imports
were significant at the level 1% and positive effect on economic growth in the Arab
countries. Also the elasticity of imports in the Arab countries during the study
period recorded 0.29 %, if the imports increased by 100% in the Arab countries the
economic growth increased by 29 per cent during the period 1995 to 2013. The
gross capital formation also was significant at level of 1% and had positive effect
on the economic growth in the Arab countries. Also the elasticity of gross capital
formation in the Arab countries during study period recorded 0.12%. It means if
gross capital formation increase by 100% the GDP in Arab countries increased by
12 percent during the period 1995 to 2013.
The labour was significant at the level 1%, and had positive on the economic
growth in the Arab countries. Also the elasticity of labour in Arab countries was
0.37% during the period 1995 to 2013. It means the labour in Arab countries
increased by 100% the GDP increased by37per cent, during the study period.

JEB, 3(1), B.S.S. Hamdan. p.100-107.

105

Journal of Economics Bibliography

6. Conclusion and Policy Recommendations
The study aimed to find estimated the effect exports and imports of the
economic growth in the Arab countries during the period 1995 to 2013, through a
form of panel data which includes economic growth measured by GDP as the
dependent variable, and a number of independent variables, which included
exports, imports, labor and Gross capital formation, in 17 Arab countries. The
countries studied were Jordan, United Arab Emirates, Bahrain, Tunisia, Algeria,
Saudi Arabia, Sudan, Oman, Qatar, Kuwait, Lebanon, Djibouti, Mauritania, Egypt,
Morocco, Yemen and Palestine. Number of countries which could have been part
of the sample was omitted due to lack of sufficient data on some of the variables
under investigation. The study the exports, imports, labor and gross capital
formation had a positive effect on economic growth in Arab countries during the
period 1995 to 2013. The study recommends the following policy measures for the
economic growth in Arab countries. As long as the gross capital formation plays a
key role in economic growth in the Arab countries, Arab countries must encourage
increase in gross capital formation, to increase its contribution to economic growth.
Support for growth-led export in Arab countries Thus effort should be direct
towards policies that will enhance economic growth such as industrialization, in
order to impact more on exports, the need to approach the Arab countries, to
economic openness to enhance the role of exports and imports in the economic
growth policy. Also Arab countries need to focus on vocational education, through
the holding of professional training courses, because of its important role in raising
the productivity of the worker in Arab countries. It is important indicator for
measuring the efficiency and effectiveness of the work element in achieving a
certain level of the output in the production process.
There is need to increase the imports of technology for increasing labor
productivity which can directly promote economic growth, and thus improve the
standards of living in the Arab countries.

JEB, 3(1), B.S.S. Hamdan. p.100-107.

106

Journal of Economics Bibliography

References
Abugamea, G. (2015). External trade and economic growth relationship under Trouble: Time series
evidence from Palestine, Swift Journal of Economics and International Finance, 1(2), 10-16.
Afaf, A.J.S.,&Hussain, M.A. (2015). Impact of exports and imports on economic growth: Evidence
from Tunisia, Journal of Emerging Trends in Economics and Management Sciences, 6(1), 13-21.
Al-Sawaee, K. (2008). Export Led Growth Hypothesis Case Study of West Asia Region, Journal of
Dirasat, Deanship of Academic Research, 35(2), 273-282.
Alexiou, C. (2001). Effective Demand and Unemployment the European Case: Evidence from
Thirteen Countries, in UNSPECIFIED, Cofru, Greece.
Chmelarova, V. (2007). The Hausman Test, and Some Alternatives with Heteroskedastic Data,
Unpublished Ph.D. theses, State University, USA.
Dinardo, J., & Johnston, J. (1997).Econometrics Methods, Fourth Edition, McGraw- Hill Companies,
Inc, New York.
Elbeydi, K.R.M., Hamuda, A.M., &Gazda,V. (2010). The Relationship between Export and
Economic Growth in Libya Arab Jamahiriya, Theoretical and Applied Economics, 1(542), 69-76
Feder, G. (1982), On export and economic growth, Journal of Development Economic, 12(1-2), 5973. Doi. 10.1016/0304-3878(83)90031-7
Green, W. (2002). Econometric Analysis, Fifth Edition, Upper Saddle River, New Jersey, Prentice
Hall, p299.
Hsiao, C. (2003). Analysis of Panel Data, Second Edition, United Kingdom the University Press,
Cambridge.
Kalaitzi, A. (2013).Exports and Economic Growth in the United Arab Emirates, Submitted to: RIBM
Doctoral Symposium, Manchester Metropolitan University Business School.
Verbeek, M. (2000).A Guide to Modern Econometrics, British Library Cataloguing in Publication
Data.

Copyrights
Copyright for this article is retained by the author(s), with first publication rights granted to
the journal. This is an open-access article distributed under the terms and conditions of the
Creative Commons Attribution license (http://creativecommons.org/licenses/by-nc/4.0).

JEB, 3(1), B.S.S. Hamdan. p.100-107.

107